AI 'intelligent surveying' transforms transmission line design

Staff from the State Grid Jiaxing Power Supply Company recently used an AI-based transmission engineering survey and design tool to complete the construction blueprint for the 110 kilovolt Yuxin 1384 / Yanyu 1385 line relocation project, significantly improving the efficiency and quality of transmission line survey and design work and shortening the overall design cycle.

As a core component of power grid construction, transmission engineering survey and design involve numerous tasks, including complex terrain analysis, environmental assessments, and engineering feasibility studies. Traditional methods are heavily reliant on manual surveying and experiential judgment, which are often hindered by challenges such as low data acquisition efficiency, limited information-processing capability, and cumbersome route-optimization processes. These shortcomings make it difficult to meet the intelligent and digital development requirements of modern power systems.

The newly deployed intelligent design tool, supported by collaboration between large models and specialized small models, integrates cutting-edge technologies such as AI-based automatic recognition, multisource data fusion, and deep learning. It establishes an intelligent design system covering the entire workflow — from data collection and route planning to drawing generation and report writing — fundamentally transforming the previous labor-intensive, experience-driven model.

The AI-powered transmission engineering survey and design tool not only greatly enhances the accuracy of terrain data extraction and improves the rationality of route planning, but also reduces manual intervention and effectively shortens the design cycle. During the data collection stage for the 110 kV Yuxin 1384 / Yanyu 1385 line relocation project, staff applied a fusion of lightweight convolutional neural networks and deep learning models to extract sub-meter-level terrain and vegetation information, enabling automated interpretation of high-resolution satellite imagery. In engineering drawing processing, after applying computer vision and natural-language integration technology, key-information recognition accuracy exceeded 96 percent.

During the route-planning phase, the system leveraged multisource data fusion to integrate GIS data, historical engineering records, and government planning information. Using reinforcement-learning algorithms, it automatically balanced constraints such as cost and environmental protection to generate optimal solutions. Additionally, through an integrated process from one-click drawing generation to intelligent report writing, AI can automatically produce construction drawings and important report content, followed by algorithm-based quality control, reducing traditional textual design time by 70 percent.

In the future, State Grid Jiaxing Power Supply Company will continue to enhance the integration accuracy of 3D models with real-world scenes, further optimize AI model performance, and expand application scenarios for intelligent route selection and object recognition. These efforts aim to accelerate transmission engineering survey and design to become smarter, more efficient, and more precise — injecting stronger technological momentum into the development of intelligent power grids.